Feyza Keceli

An Adaptive Multimedia QoS Scheduler for 802.11e Wireless LANs

The IEEE 802.11e standard defines the Hybrid Coordination Function (HCF) which specifies the QoS algorithms in the MAC layer of the next generation 802.11 wireless LAN (WLAN). The HCF is composed of two access functions: A distributed contention-based channel access function (EDCA) providing prioritized QoS and a centralized polling-based channel access function (HCCA) providing parameterized QoS. In this paper, we address the efficient medium access control (MAC) of the mostly variable bit rate (VBR) and bursty multimedia traffic in HCCA and propose a novel application-aware adaptive 802.11e QoS scheduler. The proposed scheduler coordinates the resource allocation of the associated flows using adaptive service intervals, transmit opportunities, and polling order. The service schedule is based not only on the traffic classification done according to the packet size, interval distribution, and the direction but also on the instantaneous buffered traffic conditions. The proposed scheduler is shown to have significant performance gains in terms of multimedia traffic channel utilization, packet loss ratio, delay, and jitter when compared with the other schemes.

TCP ACK Congestion Control and Filtering for Fairness Provision in the Uplink of IEEE 802.11 Infrastructure Basic Service Set

Most of the deployed IEEE 802.11 wireless local area networks (WLANs) use infrastructure basic service set (BSS) in which an access point (AP) serves as a gateway between wired and wireless domains. We present the transport layer uplink unfairness problem that occurs due to uneven bandwidth sharing between upstream transmission control protocol (TCP) data and downstream TCP acknowledgment (ACK) packets. We propose ACK congestion control and filtering in the WLAN downlink to improve fairness between uplink flows. Through extensive simulations, we show that the proposed heuristic algorithms not only enhance both short- and long-term fairness but also increase channel utilization in the 802.11 WLAN. Another attractive feature for the proposed algorithms is the independence of the performance from variable TCP round trip times (RTTs).

Saturation Throughput Analysis of the 802.11e Enhanced Distributed Channel Access Function

We propose an analytical model for the performance analysis of the enhanced distributed channel access (EDCA) function of the IEEE 802.11e standard. The proposed discrete-time Markov chain (DTMC) model incorporates the main quality-of-service (QoS) features of 802.11e, namely, contention window (CW), arbitration interframe space (AIFS), and transmit opportunity (TXOP) differentiation. Due to its specific design, the proposed DTMC model jointly considers the state of the medium access control (MAC) layer buffer and the MAC differentiation mechanisms which facilitates a novel performance analysis framework at an arbitrary traffic load. Analytical and simulation results are compared to demonstrate the accuracy of the proposed approach for varying traffic load, EDCA parameters, and MAC layer buffer space.

Weighted Fair Uplink/Downlink Access Provisioning in IEEE 802.11e WLANs

We present the unfairness problem between the uplink and the downlink flows in the IEEE 802.11e infrastructure basic service set (BSS) when the default settings of the enhanced distributed channel access (EDCA) parameters are used. First, we propose an analytical model to calculate the EDCA parameter set that ideally achieves a predetermined utilization ratio between uplink and downlink flows. Next, we design a simple and practical model-assisted measurement-based dynamic parameter adaptation algorithm. In the design of the proposed algorithm, we also consider the interactions of the bi-directional communication structure of transmission control protocol (TCP) with the medium access control (MAC) layer algorithm. Via simulations, we show that our solution provides short- and long-term fair access for uplink and downlink TCP and user datagram protocol (UDP) flows with different wired link propagation delays in a wired-wireless heterogeneous scenario. In the meantime, the quality-of-service (QoS) requirements of coexisting real-time flows can also be maintained.

Modeling the 802.11e Enhanced Distributed Channel Access Function

The Enhanced Distributed Channel Access (EDCA) function of IEEE 802.11e standard defines multiple Access Categories (AC) with AC-specific Contention Window (CW) sizes, Arbitration Interframe Space (AIFS) values, and Transmit Opportunity (TXOP) limits to support MAC-level Quality-of- Service (QoS). In this paper, we propose an analytical model for the EDCA function which incorporates an accurate CW, AIFS, and TXOP differentiation at any traffic load. The proposed model is also shown to capture the effect of MAC layer buffer size on the performance. Analytical and simulation results are compared to demonstrate the accuracy of the proposed approach for varying traffic loads, EDCA parameters, and MAC layer buffer space.

Performance Analysis of the IEEE 802.11e Enhanced Distributed Coordination Function Using Cycle Time Approach

The recently ratified IEEE 802.11e standard defines the enhanced distributed channel access (EDCA) function for quality-of-service (QoS) provisioning in the wireless local area networks (WLANs). The EDCA uses carrier sense multiple access with collision avoidance (CSMA/CA) and slotted binary exponential backoff (BEB) mechanism. We present a simple mathematical analysis framework for the EDCA function. Our analysis considers the fact that the distributed random access systems exhibit cyclic behavior where each station successfully transmits a packet in a cycle. Our analysis shows that an AC-specific cycle time exists for the EDCA function. Validating the theoretical results via simulations, we show that the proposed analysis accurately captures EDCA saturation performance in terms of average throughput, medium access delay, and packet loss ratio. The cycle time analysis is a simple and insightful substitute for previously proposed more complex EDCA models.

A capacity analysis framework for the IEEE 802.11e contention-based infrastructure basic service set

We propose a multimedia capacity analysis framework for the Enhanced Distributed Channel Access (EDCA) function of the IEEE 802.11e standard. Our analysis shows that the multimedia capacity of the EDCA function for each Access Category (AC) can accurately be estimated by appropriately weighing the service time predictions of a saturation model over different number of active stations. We propose a simple and generic cycle time model to derive the service time in saturation which we employ in the calculation of an accurate station- and AC-specific queue utilization ratio. Based on the estimated queue utilization ratio, we design a simple model-based admission control scheme. We show that the proposed call admission control algorithm maintains satisfactory user-perceived quality for coexisting voice and video connections in an infrastructure Basic Service Set (BSS) and does not present over-or under-admission problems of previously proposed models in the literature.

Achieving Fair TCP Access in the IEEE 802.11 Infrastructure Basic Service Set

We illustrate the transport layer unfairness problem in the IEEE 802.11 Wireless Local Area Networks (WLANs). We design a link layer access control block for the Access Point (AP) in order to provide fair Transmission Control Protocol (TCP) access in an 802.11 infrastructure Basic Service Set (BSS). The novel and simple idea of the proposed control block is employing a congestion control and filtering algorithm on TCP Acknowledgment (ACK) packets of uplink flows, thereby prioritizing the access of TCP data packets of downlink flows at the AP. We quantify the parameters of the proposed ACK congestion control and filtering algorithm based on the measured average downlink TCP data access rate. Via simulations, we show that the introduction of the proposed link layer access control block to the protocol stack enhances short- and long-term fairness in a wide range of scenarios.

Fair and Efficient TCP Access in IEEE 802.11 WLANs

We illustrate the transport layer unfairness problem in the IEEE 802.11 wireless local area networks (WLANs). We propose a novel and simple analytical model to calculate the per-flow congestion window limit that provides fair TCP access in a wired/wireless scenario where the wireless hop is an 802.11 link. The proposed analysis is unique in that it considers the effects of varying number of uplink and downlink TCP flows, varying round trip times (RTTs) of TCP connections, and the use of delayed TCP acknowledgment (ACK) mechanism. The comparison with simulation results validates the accuracy of the analytical estimations on the congestion window limit. When the TCP connections use the congestion window limits calculated from the proposed model, not only fair access can be provisioned, but also the channel can be utilized efficiently in a wide range of scenarios.

Multimedia Capacity Analysis of the IEEE 802.11e Contention-Based Infrastructure Basic Service Set

We propose a novel and simple multimedia capacity analysis framework for the IEEE 802.11e Enhanced Distributed Channel Access (EDCA) function. We design a simple model- based admission control scheme which is based on the accurate queue utilization ratio we calculate by appropriately weighing the service time predictions of our previously developed saturation cycle time model for different number of active stations. We show that the proposed call admission control algorithm maintains satisfactory user-perceived quality for coexisting voice and video connections in an infrastructure Basic Service Set (BSS) and does not present over- or under-admission problems of previously proposed models in the literature.